Hello, I agree with your statement that the traffic light cycle should not affect lane changing. However, it does cause issues with reproducibility as I am looking at a different situation from the one in your screenshot due to having a different phase cycle compared to to yours.
Nevertheless I have identified 2 issues in your scenario that have an impact on junction service rate: 1) Straight going vehicles are being blocked by left-turning vehicles wherever there is a common lane for left and straight movements. Unless this layout is intentional, you may want to modify the connections. The current setup also creates the risk of having straight-going vehicles block left-movers in the exclusive left-turn phase. 2) A deadlock is created due to invalid cooperative speed adjustments. ( https://github.com/eclipse/sumo/issues/4737). A work-around may be to set the attribute lcPushy > 0 for your vehicles (in particular the motorcycles). 3) right turning vehicles are sometimes blocked by straight going vehicles from the same lane (as in the deadlock above). The frequency of this happening may be reduced by setting the parameter lcTurnAlignmentDistance>0 which causes right turning vehicles to change their preferred alignment to "right" ahead of the intersection regards, Jakob Am Fr., 26. Okt. 2018 um 17:08 Uhr schrieb Khilan Ravani < [email protected]>: > Sir, > > Thank you so very much for such a detailed response to our query! > I'm very sorry, I forgot to mention this in my previous mail. Actually, > after sending the first mail, we made some changes in the parameters and > were able to partially resolve the problem. Initially, we had set the step > length value to '0.1' and left the 'actionStepLength' value to its default > i.e '1'. Then, we made some changes and set the 'actionStepLength' value to > 0.5 and the step length value also to 0.5. This seemed to partially > resolve our error. > > Following are the details to your questions : > > 1) Instead of including the lateral resolution in sumocfg file, we are > adding the script for sublane in console command itself (which makes it > something like: sumo-gui --lateral-resolution 1.75 -n project.net.xml -c > project.sumo.cfg). We are using a lateral resolution of 1.75 as we have set > the width of each lane to 3m. > > 2) Yes sir, we are using TraCI command, but just to count the number of > vehicles passing over the detectors (E1 and E3 detectors) and to change the > traffic light phases, based on those values. In my humble opinion, I don't > think that this should have any effect on lane changing. > . > 3) project.add.xml file contains the details/attributes of E1 and E3 > detectors which we have used in simulation (we have attached this file > along with this mail). > > 4) Yes sir, I'm very sorry. I forgot to mention this in the previous mail > that we changed the step length value to 0.5 and have sent you the updated > xml files. > > The problem has been partially resolved as of now. As per the current > scenario, the vehicles are changing the lanes as per their defined routes > and not completely obstructing the traffic in that direction. But, the > vehicles are taking significant amount of time to change the lanes (about > one complete cycle) which is affecting our results (Junction service rate). > > PFA a small illustration for the same (Attached images) occurring at > junction number 10. > > PFA updated files. > > Thanks again, > Sincerely, > Khilan. > > On Fri, Oct 26, 2018 at 6:43 PM Jakob Erdmann <[email protected]> > wrote: > >> Using these files I am unable to reproduce the stated problem (when >> looking at junction 9 at time 856). >> However, your input files raise some questions: >> 1) your issue is with the sublane model but your sumocfg file does not >> set the lateral resolution (I tested with a value of 0.5). What value do >> you use? >> 2) your sumocfg file uses option --remote-port. Are you sending TraCI >> commands that could have an effect on the reported lane-changing issue? >> 3) your sumocfg file references the file project.add.xml. Does this file >> contain additional settings which might affect the issue? >> 4) Your screenshot shows the time 856.2 but your configuration file sets >> a step-length of 0.5. What are the correct settings for reproducing the >> issue? >> >> regards, >> Jakob >> >> >> Am Fr., 26. Okt. 2018 um 10:19 Uhr schrieb Khilan Ravani < >> [email protected]>: >> >>> Sir, >>> >>> PFA the files for reproducing the problem. >>> >>> Thank you, >>> Sincerely, >>> Khilan. >>> >>> On Fri, Oct 26, 2018 at 12:31 AM Jakob Erdmann <[email protected]> >>> wrote: >>> >>>> Please attach minimal input files fore reproducing the problem >>>> (.net.xml, .sumocfg, .rou.xml) >>>> >>>> Am So., 21. Okt. 2018 um 16:10 Uhr schrieb Khilan Ravani < >>>> [email protected]>: >>>> >>>>> Sir, >>>>> >>>>> We are trying to use SUMO for simulating scenarios with sub-lanes. >>>>> Following is a brief description of our simulation setup (Map >>>>> attached) : We have 4 x 4 grid with 16 intersections and we have both 2 >>>>> lane and 3 lane roads in our map. Vehicles are inserted only at the end >>>>> points in edges and the routes are defined from one end point to some >>>>> another end point. Sub laning is allowed for vehicles. >>>>> >>>>> We have upgraded to version 1.0.1 and are facing some issues while >>>>> simulation. >>>>> >>>>> Following is the description of one such scenario (from the attached >>>>> image) : In a three lane road with sub-lanes, the encircled vehicle (type >>>>> = passenger) wants to take a right turn but it arrived in a lane which >>>>> does >>>>> not permit a right turn. Hence the vehicle has to first change the lane >>>>> (at >>>>> that very point) which it is unable to do, because of the traffic on the >>>>> right lane. As a consequence of this, the vehicle gets stuck at the >>>>> junction. This obstructs the following vehicles leading to a traffic jam >>>>> at >>>>> that junction. >>>>> >>>>> We believe that we might have overlooked some parameters which might >>>>> be causing this issue but are unable to figure it out. Please help us in >>>>> resolving this issue. >>>>> >>>>> Attachments : Map and scenario (with error) >>>>> >>>>> Thanks in advance! >>>>> Sincerely, >>>>> Khilan and Manish. >>>>> >>>>> _______________________________________________ >>>>> sumo-user mailing list >>>>> [email protected] >>>>> To change your delivery options, retrieve your password, or >>>>> unsubscribe from this list, visit >>>>> https://dev.eclipse.org/mailman/listinfo/sumo-user >>>>> >>>> _______________________________________________ >>>> sumo-user mailing list >>>> [email protected] >>>> To change your delivery options, retrieve your password, or unsubscribe >>>> from this list, visit >>>> https://www.eclipse.org/mailman/listinfo/sumo-user >>>> >>> _______________________________________________ >>> sumo-user mailing list >>> [email protected] >>> To change your delivery options, retrieve your password, or unsubscribe >>> from this list, visit >>> https://www.eclipse.org/mailman/listinfo/sumo-user >>> >> _______________________________________________ >> sumo-user mailing list >> [email protected] >> To change your delivery options, retrieve your password, or unsubscribe >> from this list, visit >> https://www.eclipse.org/mailman/listinfo/sumo-user >> > _______________________________________________ > sumo-user mailing list > [email protected] > To change your delivery options, retrieve your password, or unsubscribe > from this list, visit > https://www.eclipse.org/mailman/listinfo/sumo-user >
_______________________________________________ sumo-user mailing list [email protected] To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user
